Dendritic cells (DCs) are highly specialized antigen presenting cells found in all lymphoid and non-lymphoid tissues. In the traditional view, DCs are terminally differentiated cells constantly replaced by a pool of circulating precursors. In contrast to this view, we recently demonstrated that in mice, epidermal DCs also called Langerhans cells (LCs) are maintained by local radio-resistant proliferative precursors, throughout life in quiescent skin, and are replaced by circulating precursors only during injuries. We have also discovered that host radio-resistant LCs can persist in recipient mice after allogeneic bone marrow transplant (allo-BMT). Graft versus host disease (GVHD) is a major cause of mortality and morbidity after allo-BMT. Importantly, we found that persistence of host LCs after allo-BMT correlates with severe skin GVHD, while replacement of host LCs by donor LCs improves skin GVHD outcome. Therefore, exploring DC homeostasis in a tissue target of GVHD may help develop new strategies for the treatment of this devastating disease. In addition to LCs, we recently discovered that a subset of DCs in the gut can resist lethal doses of radiation, raising the possibility that LCs may not be unique in being radio-resistant;other tissues may harbor radio- resistant populations that contribute to GVHD. Thus, in aim 1, we will examine the turnover of DCs in the gut, a tissue that is particularly affected by GVHD. Gut DCs constantly transport commensal bacteria and pathogens from the luminal intestine to local lymphoid tissues, playing a critical role in innate and adaptive immunity. To face their continuous migration, gut DCs must be replaced with new cells, but the nature of the committed precursor that maintains the gut DC pool remains elusive. Using novel strategies, we have recently characterized the circulating LC precursor as being a specific subset of circulating monocytes. We also established the sequence of events that leads a monocyte to become a LC and discovered that signaling through the receptor for CSF-1 is required for this process (Nature Immunology in press). Therefore in aim 2, we propose to use similar approaches to explore DC homeostasis in the gut in the steady state and identify the circulating precursor that gives rise to intestinal DCs. Finally, our finding that induction of DC chimerism is critical to prevent GVHD leads us, in aim 3, to determine the nature of the circulating precursor that repopulates DCs in a clinically relevant model for allo-BMT.